GitLab CI transient failures
Robert Goldman <[email protected]> Tue, 01 Jul 2025 10:13:53 -0500
| Newsgroups | gmane.lisp.common-lisp-net.devel |
|---|---|
| Message-ID | <[email protected]> |
--=_MailMate_E1E18F19-8DAE-41B1-8A19-F9AAE578FAC1_= Content-Type: text/plain; format=flowed; markup=markdown I really have no idea why this is happening, but on the ITERATE GitLab repo, I am getting those `shell not found` errors again. And what's really weird is that this is transient. After a failure I can re-run the tests and sometimes they will pass. This is all quite confusing. In the past, I have seen internet posts suggesting an issue with the GitLab conf that can be fixed by resetting the entry for the shell, which apparently tries to guess what shell to use based on the image. I don't see how that's compatible with this failure being transient, though. Best, R --=_MailMate_E1E18F19-8DAE-41B1-8A19-F9AAE578FAC1_= Content-Type: text/html Content-Transfer-Encoding: quoted-printable <!DOCTYPE html> <html> <head> <meta http-equiv=3D"Content-Type" content=3D"text/xhtml; charset=3Dutf-8"= > </head> <body><div style=3D"font-family: sans-serif;"><div class=3D"markdown" sty= le=3D"white-space: normal;"> <p dir=3D"auto">I really have no idea why this is happening, but on the I= TERATE GitLab repo, I am getting those <code style=3D"margin: 0 0; paddin= g: 0 0.25em; border-radius: 3px; background-color: #F7F7F7;">shell not fo= und</code> errors again. And what's really weird is that this is transien= t. After a failure I can re-run the tests and sometimes they will pass. T= his is all quite confusing. In the past, I have seen internet posts sugge= sting an issue with the GitLab conf that can be fixed by resetting the en= try for the shell, which apparently tries to guess what shell to use base= d on the image. I don't see how that's compatible with this failure being= transient, though.</p> <p dir=3D"auto">Best,<br> R</p> </div> </div> </body> </html> --=_MailMate_E1E18F19-8DAE-41B1-8A19-F9AAE578FAC1_=--